A Study On Drought Monitoring By Using Multiple Approachs

Food security is the basic key to ensure the steady development of the economy and society. According to the historical data, in many natural disasters, drought has been the most serious disaster which troubled the national agricultural production. In recent decades, the occurrence frequency of drought significantly increased, so it is necessary to study the changes of the drought, especially in the context of the global climate change. This study mainly focused on drought parameter retrieval and drought monitoring method evaluation and the drought variations in the major crop production regions in China.First, former research achievements focused on drought and several drought monitoring methods were shown briefly, and comparisons of several drought monitoring indexes indicated that the precision of the meteorological drought monitoring methods was higher than remote sensing-base methods which can obtain the drought spatial distribution more easily. Among the six kinds of the drought monitoring methods, only the monitoring results of SPI and SPEI indexes were comparable in different regions, and the results of remote sensing-base methods usually showed a long time delay.Second, comparisons were made between the Visible Infrared Imaging Radiometer Suite (VIIRS) which is the NASA’s next generation earth observation sensor and the Moderate Resolution Imaging Spectroradiometer (MODIS), and it indicated that VIIRS performed better than MODIS. Then, a practical algorithm was proposed to retrieve land surface temperature (LST) from VIIRS data in mid-latitude region which overcame the shortage that VIIRS did not owns the vapor channel, and the cloud detection algorithms were given among China region. The validation with the standard atmospheric simulation showed the average LST retrieval error for the algorithm was not great than 1.0 K, and the estimation accuracy for the cloud test methods was higher than 85%, which meant the LST and cloud detection algorithms were available for the agricultural drought monitoring.Third, the suitability of six kinds of drought monitoring index was evalulated in China region by using three sites. It indicated that Palmer index with abnormal conversion of dry and wet, and incomparability at different sites, performed worse than SPEI and SPI. The performences of the SPEI and SPI were similar, but the SPI could obtain more accurate drought information. Three kinds of remote sensing monitoring indexes of drought monitoring had a long time delay, the monitoring results in different regions could be compared well, but the NDVI-Ts index performed generally better than SVI index and VCI index. The SPI index based on TRMM data was similar to the SPI index which could monitor the drought accurately, and the more reasonable time scale for the summer drought monitoring in the South was 7 days.Finally, this study analyzed the occurrence of droughts and the patterns of drought changes over the past 60 years in China’s major crop-growing areas based on the long time sequenced precipitation data. Regarding the temporal distribution of severe drought in major crop-growing areas, the most serious drought of the Southwest China area and the Sichuan area occurred in the early 21st century, the drought occurrence during 1960 to 1970 in the middle and lower reaches of the Yangtze River (MLRY) and South China area was most severe, and around 2005 was the most severe drought year in the Northeast China area and Huang-Huai-Hai area. Regarding the trend of drought occurrences, the southern areas had become dryer in recent years, whereas the northern areas were currently in a dry-to-wet transition. Particularly for the southern areas, the Southwest China area and Sichuan may experience less occurrence of drought, while the MLRY area and South China area may experience increased aridity. The analysis of the spatial distribution of droughts in the past 15 years by using TMI rainfall data revealed that the north-to-south or south-to-north shift of aridity in China had a cycle of 15 to 20 years. Spatially, the pattern of "southern flood, northern drought; northern flood, southern drought" was apparent.